The trypanocidal activity of the ODC (ornithine decarboxylase) inhibitor DFMO (difluoromethylornithine) has validated polyamine biosynthesis as a target for chemotherapy. As DFMO is one of only two drugs used to treat patients with late-stage African trypanosomiasis, the requirement for additional drug targets is paramount. Here, we report the biochemical properties of TbSpSyn (Trypanosoma brucei spermidine synthase), the enzyme immediately downstream of ODC in this pathway. Recombinant TbSpSyn was purified and shown to catalyse the formation of spermidine from putrescine and dcSAM (decarboxylated S-adenosylmethionine). To determine the functional importance of TbSpSyn in BSF (bloodstream form) parasites, we used a tetracycline-inducible RNAi (RNA interference) system. Down-regulation of the corresponding mRNA correlated with a decrease in intracellular spermidine and cessation of growth. This phenotype could be complemented by expressing the SpSyn (spermidine synthase) gene from Leishmania major in cells undergoing RNAi, but could not be rescued by addition of spermidine to the medium due to the lack of a spermidine uptake capacity. These results therefore genetically validate TbSpSyn as a target for drug development and indicate that in the absence of a functional biosynthetic pathway, BSF T. brucei cannot scavenge sufficient spermidine from their environment to meet growth requirements.
As a result of co-ordinated surveillance and treatment programmes, the number of people infected annually by the protozoan parasite Trypanosoma brucei, the causative agent of HAT (human African trypanosomiasis), has fallen to 70000 . However, in some regions, mortality rates for HAT exceed HIV/AIDS and malaria. Current treatments for the late stage of the disease rely on the drugs melarsoprol and DFMO (difluoromethylornithine). However, their use is problematic. With no realistic prospect of a vaccine, new drugs against ‘one of the great neglected diseases’ are a priority .
One area of parasite biology that has attracted attention in terms of drug development is polyamine biosynthesis [3–6]. Polyamines such as putrescine and spermidine are aliphatic organic compounds containing two or more amino groups. In most eukaryotes, these molecules play pivotal roles in several processes, including protein/nucleic acid synthesis and cell proliferation/differentiation, and have been implicated in the development of certain cancers . Trypanosomes are unusual in that spermidine is also used in the synthesis of trypanothione, a glutathione–spermidine dithiol conjugate that plays a central role in several detoxification processes and supplies reducing equivalents used in nucleic acid synthesis [8,9]. As this thiol is both unique and essential to trypanosomatids, any process involving trypanothione is considered a potential target for drug intervention.
The building blocks for polyamine synthesis are arginine and methionine (Figure 1). In most eukaryotes, arginine is hydrolysed to form ornithine and urea, with ornithine metabolized further to putrescine, reactions catalysed by arginase and ODC (ornithine decarboxylase) respectively. Putrescine then reacts with dcSAM (decarboxylated S-adenosylmethionine) to form spermidine, a process mediated by SpSyn (spermidine synthase); the dcSAM is derived from methionine as a result of the activity of SAMdc (S-adenosylmethionine decarboxylase) (Figure 1). The trypanocidal activity of various inhibitors that target components of this pathway has been reported [10–14]. DFMO, one of two drugs available to treat late-stage HAT, inhibits ODC activity. The interaction between T. brucei ODC and DFMO has been extensively studied and the selective killing activity of the drug has been determined [15,16]. The basis for this toxicity stems from the stability of the parasite ODC (half-life ∼18 h) compared with that of the mammalian host ODC (half-life ∼20 min) [15,17]. When DFMO binds covalently to the mammalian enzyme, the complex is rapidly degraded and replaced with newly synthesized ODC. In contrast, the DFMO-ODC molecule remains within the parasite and is only slowly replaced. Thus the level of active protein is decreased leading to a cessation of putrescine formation. Eventually, the parasite stops growing and the non-dividing cells are cleared by the host immune system. Although the activities of other T. brucei enzymes involved in polyamine synthesis have been detected [12,18], their biochemical roles have yet to be fully analysed. Here, using an inducible RNAi (RNA interference) system, we demonstrate that SpSyn (EC 184.108.40.206), a key component of polyamine synthesis, is essential to infective T. brucei, therefore validating it as a potential target for chemotherapeutic intervention.
Polyamine biosynthetic pathway in trypanosomatids
BSFs (bloodstream forms) of the SMB (T. brucei single marker cell line) that constitutively express T7 RNA polymerase and the tetracycline repressor protein  were grown at 37 °C under a 5% CO2 atmosphere in modified Iscove's medium  containing 2 μg·ml−1 G418. Transformed SMB cells were maintained in this growth medium, supplemented with 2.5 μg·ml−1 hygromycin. Tetracycline-free fetal calf serum (10%, v/v) (Autogen Bioclear) was used in the growth medium. RNA was extracted from parasites by using the RNeasy® mini kit (Qiagen).
TbSpSyn (Trypanosoma brucei spermidine synthase) was amplified from parasite genomic DNA (see Table 1 for primers). The fragment was digested with BamHI+HindIII and cloned into the corresponding sites of the expression vector pTrcHis-C (Invitrogen). In this system, the expressed protein contains an N-terminal His tag and an epitope detectable with the anti-Xpress monoclonal antibody (Invitrogen). The insert was sequenced using a dye terminator cycle kit (Applied Biosystems) and an ABI Prism 3730 sequencer. Escherichia coli BL21+ transformed with pTrcHis-TbSpSyn were grown in NZCYM broth (Sigma) containing ampicillin, and protein expression was induced by the addition of IPTG (isopropyl β-D-thiogalactoside). Recombinant His-tagged proteins were affinity-purified on an Ni-NTA (Ni2+-nitrilotriacetate) matrix column under native conditions as recommended by the manufacturer (Qiagen). The cell lysis, column wash and elution steps were all carried out in the presence of protease inhibitors (Roche). Fractions were analysed by SDS/PAGE and protein concentrations were determined by a BCA (bicinchoninic acid) protein assay system (Pierce).
TbSpSyn activity was followed by monitoring spermidine formation in the presence of putrescine and dcSAM (modified from ). A standard 1 ml reaction contained 100 mM potassium phosphate (pH 7.2), 1 mM dithiothreitol, 1 mM putrescine, 10 μM dcSAM and 100 μg of TbSpSyn. Reactions were incubated at 37 °C for 30 min and then terminated by placing the test tubes at 95 °C for 5 min. Polyamines were derivatized with dansyl chloride and then analysed by HPLC.
An 860 bp DNA fragment corresponding to an internal sequence from TbSpSyn was amplified from genomic DNA (see Table 1 for primers), digested with BamHI+XhoI and cloned into the corresponding sites of the vector p2T7Ti . In this vector, the inserted DNA is flanked by two opposing T7 promoters, with each promoter under the control of a tetracycline operator. Constructs were linearized with NotI and electroporated into BSF parasites and transformants were cloned as described . Induction of RNAi was initiated by adding tetracycline (1 μg·ml−1) to the culture.
The complete ORF (open reading frame) of LmSpSyn (Leishmania major SpSyn) (CAC44919) was amplified from L. major genomic DNA (see Table 1 for primers), and the resultant 903 bp fragment was digested with SbfI+AscI and then cloned into the corresponding sites of the T. brucei constitutive expression vector pTUB-EX . Constructs were digested with NotI+XhoI and electroporated into the RNAi-TbSpSyn cell line and clones were selected. Comparison between TbSpSyn and LmSpSyn revealed that the two genes were 65% identical and had no conserved stretches >20 bp. Therefore RNAi targeted at the TbSpSyn mRNA should not affect the Leishmania transcript.
Detection of the growth inhibition phenotype
BSF trypanosomes, transformed with the RNAi construct, were seeded at 1×105 cells·ml−1 and incubated at 37 °C in the presence of tetracycline (1 μg·ml−1). Every 24 h, parasite growth was monitored microscopically and the culture was diluted back to 1×105 cells·ml−1. Control cultures incubated in the absence of tetracycline were grown in parallel. Under these conditions, untreated cell lines grew with a doubling time of approx. 7–8 h.
Analysis of polyamines
BSF trypanosomes (5×107) were lysed in 10% (v/v) trichloroacetic acid and sonicated and the extract was clarified. Polyamines in the lysate were then derivatized as follows: supernatant (100 μl) was incubated with 200 μl saturated sodium carbonate and 400 μl of dansyl chloride (10 mg·ml−1 in acetone) (Sigma) at 60 °C for 1 h in the dark. Excess dansyl chloride was removed by addition of 100 μl of L-proline (100 mg·ml−1) and incubated at 60 °C for a further 30 min. Derivatized polyamines were then extracted into toluene and the organic phase was dried under argon. Pellets were resuspended in acetonitrile and analysed by HPLC.
Spermidine transport studies
[14C]Spermidine trihydrochloride was supplied by GE Healthcare U.K. BSF T. brucei or Trypanosoma cruzi epimastigotes, in the exponential phase of growth, were pelleted, washed three times with assay buffer (90 mM Tris/HCl, pH 7.5, 3.1 mM KCl, 96.9 mM NaCl, 5 mM MgCl2, 2 mM Na2HPO4 and 2 mM glycerol) and then resuspended at 2×108 cells·ml−1. To initiate uptake, radiolabelled spermidine (0.5 μCi) was added to 50 μl of parasite suspension at a final concentration of 45 μM and the reaction was incubated at 22 °C for an appropriate time. The assay was stopped by the addition of 200 μl of 4% (w/v) paraformaldehyde in PBS. The cells were pelleted, washed three times with 1×PBS and then lysed in 250 μl of 2% (w/v) SDS in a scintillation vial. A 3 ml portion of scintillation fluid (Ecoscint A; National Diagnostics) was added to vials, which were incubated overnight. The incorporated radioactivity was then measured using a Beckman liquid-scintillation spectrometer. As a control, paraformaldehyde was added to trypanosomes before adding the radiolabelled spermidine. Each time point was performed in triplicate.
All HPLC equipment and software were from Dionex. Separations were carried out utilizing a Genesis AQ 4 μm column (150 mm×4.6 mm; GraceVydac) by eluting with 20 mM ammonium formate (pH 2.7)/acetonitrile (15:85) passing through a fluorescence detector (Dionex model RF 2000), with a λex of 340 nm and a λem of 510 nm, at a flow rate of 1.0 ml·ml−1. Peak identity was confirmed by measuring the retention time and comparison with commercially available spermidine or putrescine. A calibration curve of spermidine or putrescine was generated with Chromeleon (Dionex software) by using known amounts of the standard (0–1000 μM) in acetonitrile.
Isolation of a TbSpSyn gene
A DNA sequence (GeneDB accession no. XM_822031) in the T. brucei database was identified as containing an ORF related to spermine/spermidine synthases (Pfam PF01564). Based on this, we amplified an 894 bp DNA fragment with potential to encode a 33 kDa protein (designated TbSpSyn). Pairwise alignments revealed that TbSpSyn had 39–43% sequence identity to SpSyns of mammalian, plant and fungal origin (Figure 2). The similarity extended across the entire length of the trypanosomal protein and included the PAPT (putrescine aminopropyltransferase) family signature motif [(V/A/I)-(V/A/L)-(L/I/V)-(L/I/V)-G-G-G-X-(G/C)-X-X-(L/I/V/A)-X-E] (Prosite PS01330).
Sequence analysis of TbSpSyn
To investigate the activity of the putative SpSyn, TbSpSyn was cloned into pTrcHis-C and expressed in E. coli (see the Experimental section). After induction with IPTG, a 37 kDa band was detected in the soluble fraction of bacterial extracts that could be readily purified by one round of affinity chromatography (Figure 3A). Enzyme activity was assayed by monitoring the production of spermidine from putrescine and dcSAM. After performing each reaction, the total polyamine content was derivatized with dansyl chloride and then analysed by HPLC coupled with a fluorescence spectrophotometer. Standard curves were generated by plotting different [spermidine] against the HPLC-derived area under the peak. From this, the polyamine levels in each assay could be determined. Initial reactions using 1 mM putrescine and 10 μM dcSAM in the presence of recombinant TbSpSyn demonstrated that the parasite enzyme could mediate the formation of spermidine. To investigate further these interactions, assays were carried out using various concentrations of putrescine (12.5–1000 μM) at a fixed concentration of dcSAM (10 μM). Double reciprocal plots of 1/TbSpSyn activity against 1/[putrescine] were linear (Figure 3B) and extrapolation allowed the Km (app) value for putrescine (205±65 μM) and Vmax (11.9±1.4 nmol of spermidine formed·min−1·mg−1) to be calculated. To determine whether TbSpSyn activity could be saturated by dcSAM, reactions were carried out using a fixed concentration of putrescine (1000 μM) and variable concentrations of dcSAM (0.025–10 μM) (Figure 3C). Again, TbSpSyn demonstrated Michaelis–Menten kinetics with the parasite enzyme having a Km (app) for dcSAM of 0.09±0.01 μM.
Biochemical properties of TbSpSyn
Investigating the functional significance of TbSpSyn using RNAi
Using a tetracycline-inducible RNAi system, we set out to determine whether TbSpSyn was required for parasite viability. A DNA fragment (860 bp) was generated using PCR and cloned into p2T7Ti , and the RNAi construct was transformed into BSF T. brucei. To examine whether down-regulation of TbSpSyn affected the growth of BSF T. brucei, the cumulative cell density of tetracycline-treated parasites was compared with untreated cultures (Figure 4A). In the absence of tetracycline, the RNAi cell lines grew at the same rate as parental cultures. Addition of tetracycline had a dramatic effect (Figure 4A). Within 72 h, there was a significant decrease in the growth rate, with a reduction in the endogenous TbSpSyn mRNA (Figure 4B) and a 60% decline in spermidine levels (Table 2). Over the next 48 h, the cell density remained static. Similar results were observed with two independent clones.
Spermidine biosynthesis is essential for BSF T. brucei
|[Polyamine] (nmol/108 cells)|
|Cell line||Putrescine||Spermidine||Putrescine/spermidine ratio|
|RNAi cell line|
|[Polyamine] (nmol/108 cells)|
|Cell line||Putrescine||Spermidine||Putrescine/spermidine ratio|
|RNAi cell line|
To demonstrate conclusively that the observed growth deficiency was due specifically to a reduction in the TbSpSyn transcript, a complementation strategy was used. This involved inserting the LmSpSyn gene into the tubulin locus of the T. brucei RNAi cell line using the constitutive expression vector pTUB-EX . In these experiments, the parasites induced to undergo down-regulation of TbSpSyn while expressing LmSpSyn grew at a rate comparable with non-induced controls (Figure 4A). RNA hybridization confirmed that the level of endogenous TbSpSyn had fallen and that expression of the complementing LmSpSyn mRNA was occurring (Figure 4B).
Spermidine uptake by BSF T. brucei
Attempts to complement the RNAi-mediated growth defect by adding spermidine to the medium, an approach that has proved successful in analysing Leishmania polyamine biosynthesis-null mutants [25–28], failed. Interestingly, exogenous spermidine displayed significant trypanocidal activity at concentrations above 10 μM (results not shown). This had been previously noted and attributed to polyamine oxidase activity in the heat-inactivated fetal calf serum that is used to supplement the parasite growth medium [29,30]. At concentrations where no significant toxicity was observed (<5 μM), addition of spermidine to the medium failed to complement the RNAi-mediated growth defect, even though this level is >15-fold higher than that found in blood plasma (300 nM) . One possible explanation for this may be that T. brucei lack a high-affinity spermidine transporter related to that in L. major and T. cruzi [32,33]. To determine whether BSF T. brucei can scavenge spermidine from its environment, uptake assays were performed using radiolabelled polyamine (see the Experimental section). In these experiments, T. brucei displayed no capacity to take up spermidine, even when assays were carried out for 1 h (Figure 5). In parallel assays, T. cruzi was shown to transport spermidine. Therefore BSF T. brucei is incapable of scavenging spermidine and, as such, must rely on de novo synthesis to acquire this polyamine.
BSF T. brucei lack a spermidine uptake capacity
Polyamines are ubiquitous in nature and play crucial roles in several biological processes. All eukaryotes and eubacteria have evolved uptake and/or biosynthetic strategies to acquire them and it has been proposed that inhibitors targeting these systems may be of importance in treating infectious diseases and certain cancers [3–7]. Here, we demonstrate that SpSyn in BSF T. brucei represents a valid target for chemotherapy.
To confirm that the genome database annotation was correct, we performed SpSyn assays using purified protein generated by expression in E. coli. These experiments demonstrated that the recombinant enzyme had Km (app) values for putrescine and dcSAM of 205 and 0.09 μM respectively (Figure 3). An earlier study investigating the biochemical properties of TbSpSyn, partially purified from the parasite, noted that this enzyme functioned as a homodimer of 74 kDa, with Km (app) values for putrescine and dcSAM similar to our data . It was also observed that activity could be markedly reduced by dicyclohexylamine and cyclohexylamine, compounds that act as competitive inhibitors with respect to putrescine. In many respects, these biochemical properties are typical of PAPTs from other sources (see http://www.brenda.uni-koeln.de). For example, the Km for putrescine of HsSpSyn (human SpSyn) is 80 μM . However, TbSpSyn does appear to have a higher affinity for dcSAM compared with the human enzyme, with HsSpSyn exhibiting a Km for dcSAM (7 μM) 70-fold higher than its trypanosomal counterpart . This difference may represent a property exploitable in terms of inhibitor design.
Different trypanosomatid parasites use various mechanisms to acquire polyamines. Leishmania, for example, possess a polyamine biosynthetic pathway, plus putrescine/spermidine transporters [25–28,32], whereas T. cruzi contain a partial biosynthetic pathway and rely on uptake to meet their polyamine requirement [35–37]. To determine the importance of SpSyn to BSF T. brucei and investigate whether a biosynthetic and/or transporter mechanism is used to acquire spermidine, we used RNAi to down-regulate expression of TbSpSyn (Figures 4A and 4B). This experiment clearly demonstrated that TbSpSyn is essential for the growth of BSF T. brucei and that expression of the LmSpSyn gene in these cells can complement the resulting growth-deficient phenotype (Figure 4A). We were unable to complement the RNAi-mediated growth defect by addition of spermidine to the medium, suggesting that for the T. brucei subspecies used here, T. brucei brucei, scavenging alone does not fulfil the parasite's requirements for spermidine. One possible explanation may be the lack of a high-affinity putrescine/spermidine transporter related to that reported in L. major and T. cruzi [32,33]. Indeed, comparison of the syntenic regions of the trypanosomatid genomes failed to identify the gene encoding this carrier protein in the T. brucei brucei database. If this protein is absent, then, based on ODC gene deletion experiments, putrescine uptake must occur through other carrier(s) [38–40]. The absence of an effective spermidine permease in T. brucei, as demonstrated here (Figure 5), may reflect an adaptation to the host bloodstream environment where the plasma concentration of spermidine is low (∼300 nM) compared with the levels detected in mammalian cells (mM) where intracellular parasites such as Leishmania and T. cruzi reside [31,41].
The regulation of the intracellular polyamine pools in many eukaryotes is accomplished by a series of complex regulatory and degradation pathways with disruption of these systems promoting unwanted side-effects such as uncontrolled cell proliferation or apoptosis [42,43]. Analysis of polyamine levels in trypanosomes undergoing RNAi-mediated depletion of TbSpSyn revealed that these cells had decreased spermidine content but, unexpectedly, there had been no significant build up of putrescine (Table 2). Overexpression studies of ODC, SpSyn and SAMdc in Leishmania donovani also showed that even when enzyme activities were >25-fold higher than wild-type, there was little change in the polyamine content . This strongly implies that trypanosomatids have the ability to control metabolic flux through the polyamine biosynthetic pathway. Considering that the parasite enzymes are stable, compared with their mammalian counterparts, and that regulation of gene expression in trypanosomatids does not generally occur at the transcriptional level [15,45], the processes involved in this regulatory system may be unusual. Recently, a novel mechanism for regulating SAMdc activity in T. brucei has been reported [46,47]. An inactive version of SAMdc, known as the prozyme, forms a heterodimeric complex with the functional enzyme, which stimulates activity by >3 orders of magnitude. It will be of interest to determine whether any other enzyme in the trypanosomatid polyamine biosynthetic pathway displays an unusual regulatory mechanism.
The success of DFMO, one of two drugs used to treat late-stage HAT, has already shown that compounds targeting polyamine synthesis can be used to treat this disease. The present study demonstrates that TbSpSyn, the enzyme immediately downstream of ODC in this pathway, is essential in BSF T. brucei. This, coupled with the parasite's inability to scavenge sufficient spermidine from its environment, validates TbSpSyn as a potential target for future drug development.
We thank David Horn [LSHTM (London School of Hygiene and Tropical Medicine)] for a critical reading of this paper, Astrid Wingler (University College London) for advice on derivatization of polyamines and the Gates Malaria Partnership at the LSHTM for providing support for the analytical facility through an award from the Bill and Melinda Gates Foundation. T. brucei sequence data were obtained from The Wellcome Trust Sanger Institute website at http://www.sanger.ac.uk/Projects/T_brucei/. Sequencing of the T. brucei genome was accomplished as part of the Trypanosoma Genome Network with support by The Wellcome Trust. This work was supported by the Wellcome Trust.
human African trypanosomiasis
Leishmania major SpSyn
open reading frame
Trypanosoma brucei single marker cell line
Trypanosoma brucei spermidine synthase
Present address: School of Biological and Chemical Sciences, Queen Mary University of London, London E1 4NS, U.K.